Improving The Java Core Library

dautelle writes "Many Java developers are frustrated by the not-so-open process to improve/correct/augment the Java core libraries. Unless you work for Sun or belong to a JSR expert group, there is very little you can do to influence the future of the Java platform. Even the JCP route can be a frustrating one (e.g. JSR-108 withdrawn by Sun because not enough progress made in a timely manner). To address this serious issue, the charter of the Java Addition to Default Environment (JADE for short) has been extended, along with the release of JADE 7.0. Participation to the jade.* package development is truly open (unlike javax.*). The library already provides numerous useful classes, bug fixes, enhanced implementations of existing classes, etc. Hopefully in the near future, the library could become so useful that it becomes a de-facto complement to the JDK."

Interestin library ... this jade ... but ...

[moro_666]

i have been programming around with java for quite a long time, sometimes indeed i have felt like some c/c++ solutions would be so easy'n'fast'n'economic at some points, but i always have found a working workaround. seems like the jade dudes are not satisfied with that.

the thing that disturbs me about this article is that, it seems like the author would like to have this lib in java. i'm against it. because it provides tools to improve your code in many ways, but at the same time, it somewhat brings java down to the c++ like level, where memory losses and "forgotten" objects are quite common mistakes.

as a separate library, i think jade is great and the next time i'm writing something really complicated i'll surely have a look at it. but at the same time i think it should not be included in java's original libraries cause java newbies would surely make a lot of mistakes by using it and then everyone would blame Java for being so buggy and unusable for them. java has a strong and stable baselib which doesn't provide many ways to make mistakes, jade surely has greater opportunities but also greater flawsources.

however i think that sun should somehow support jade's development and offer at least a link to it under it's downloads page, so people could see that the usual oop model is not the limit and also would avoid inventing the wheel in a lot of cases.

i think sun is being conservative, cause it has always been that way, and it seems a quite secure and reasonable, cause most java apps are more secure and stable than anything else around. they are just concerned about java's reputation and dont want to rush items into their language which would overhaul new java developers and lead to popularity loss of java.

many cool libraries could be real battleaxes in the hands of java, but at the sametime, they could backfire, which sun is just trying to avoid.

keep up the good work on jade :)

Re:Interestin library ... this jade ... but ...

[jaoswald]

Garbage collection means you don't have to use tools that run your program in a sandbox to avoid programming errors which are unfortunately easy to make. Instead, you can spend the time making enhancements and improving performance instead of plugging stupid leaks all day.

Valgrind only fixes memory allocation problems which you can reproduce in a test situation. It doesn't fix memory allocation problems which occur when your application reaches the field and can be subjected to long runs with unpredictable inputs.

Garbage collection IS a modern programming tool. Why are you so afraid of it?

Re:Interestin library ... this jade ... but ...

[dkf]

Valgrind doesn't solve all problems. It doesn't even find all problems (unless you're extremely good at test suite authoring). It's a really cool tool, but let's not oversell it.

xml parser

[raffe]

Has anybody worked with the jade xml parser. Can someone comment? It looks interesting...

New Jade Version Released

[tod_miller]

No need to impose any library additions onto the core libraries of Java, or gripe at the JCP, however flawed, I have yet to see a similar process that works.

W3C use similar methods to develop the web standards we use every day.

Jade is a useful and in particular thier XML parsing libraries are interesting.

Look deeper and you see some interesting components. I am a little perplexed at the underlying ethos of trying to patch the core libraries with this library though.

I think the whole of Jade should be living in commons.apache.org somewhere, there is an example in invaluable libraries that I take for granted every day. That doesn't mean another programmer does, or that they should be shipped by default.

Kudos on the new version! If I ever need it, I will surely be grateful!

Re:New Jade Version Released

[sr180]

I think the whole of Jade should be living in commons.apache.org somewhere

Apache Commons has been closed. Nothing lives there now.

Re:New Jade Version Released

[Anonymous Coward]

The author of the post was referring to the http://commons.apache.org/ commons project being closed. This is not the jakarta commons at http://jakarta.apache.org/commons/index.html.

jakarta.apache.org/commons/

[trafik]

The project you seek can be found here:

http://jakarta.apache.org/commons/

Re:New Jade Version Released

[Zach Garner]

I think the whole of Jade should be living in commons.apache.org somewhere

You mean: http://jakarta.apache.org/commons/ [apache.org]

The Jakarta project is Apache's Java efforts. commons.apache.org used to hold common libraries such as APR for Apache HTTPD. These were mostly C libraries, I believe.

Apache Jakarta Commons (ok, so Apache needs to clean up and simplify there project namespace), rocks.

Here's there summary for commons-collections
* Bag interface for collections that have a number of copies of each object
* Buffer interface for collections that have a well defined removal order, like FIFOs
* BidiMap interface for maps that can be looked up from value to key as well and key to value
* MapIterator interface to provide simple and quick iteration over maps
* Type checking decorators to ensure that only instances of a certain type can be added
* Transforming decorators that alter each object as it is added to the collection
* Composite collections that make multiple collections look like one
* Ordered maps and sets that retain the order elements are added in, including an LRU based map
* Identity map that compares objects based on their identity (==) instead of the equals method
* Reference map that allows keys and/or values to be garbage collected under close control
* Many comparator implementations
* Many iterator implementations
* Adapter classes from array and enumerations to collections
* Utilities to test or create typical set-theory properties of collections such as union, intersection, and closure

For those doing Swing programming, also check out Java Desktop Network Components (JDNC [java.net]) project (this isn't from Apache, unfortunately). The documentation isn't that great yet, but the API [javadesktop.org] is all you need.

Re:New Jade Version Released

[notyou2]

I'm sure the commons-collection is a nice implementation, but for those who haven't heard, Java 1.5 (currently in beta) contains most [sun.com] of the features you've named. Some are simply new classes in the collections framework, others are language extensions (i.e. generics for type-safe collections).

Also, some of the features you named are already in java 1.4, such as identity maps [sun.com], reference maps that allow garbage collection [sun.com], adapters for converting collections to/from enumerations [sun.com] and arrays [sun.com].

An alternative ...

[LizardKing]

As an alternative to trying to contribute to the Sun java and javax libraries, people could contribute to the Classpath libraries [gnu.org]. Getting these complete means effort could then be expended on useful extensions and perhaps some optional improvements to the standard libraries. If Classpath could get some serious impetus (from IBM for instance), then Sun may have to open the development of their reference implementation or risk being left behind.

Re:An alternative ...

[MemoryDragon]

I dont think the classpath project would be a good idea, they want to emulate the core libs as good as possible. The best point to donate extensions to, is the jakarta commons project which already has many useful extensions and already has become a second core lib to many developers.

I'm confused

[Anonymous Brave Guy]

When I read the intro, I was expecting something like a Java version of C++'s Boost libraries -- things the standard library missed or didn't do well, peer-reviewed to keep the quality up, etc.

When I read the linked page, I found more like a Java version of C++ -- it looks as though a lot of those libraries are there to overcome the very strengths/weaknesses (depending on your application) that most differentiate the two languages. If you want to use C++, why not just use C++?

Alternative to the core libs

[MemoryDragon]

Donate your solutions to the Jakarta project, namely to the Apache commons project, which is basically already a second inofficial core library, used by most programs out there. I dont think you really need a revamp of the core libary addition process, Sun is doing a fine job there, since there are so many alternatives out there.

Re:I'm confused

[Glock27]

Unfortunately I missed this topic when it originally came up, but this thread seemed a good place to throw in my (belated) two cents...

When I read the linked page, I found more like a Java version of C++ -- it looks as though a lot of those libraries are there to overcome the very strengths/weaknesses (depending on your application) that most differentiate the two languages. If you want to use C++, why not just use C++?

I don't see why you think JADE is a "Java version of C++". If you're referring to the

How times change...

[crazy blade]

Many Java developers are frustrated by the not-so-open process to improve/correct/augment the Java core libraries. Unless you work for Sun or belong to a JSR expert group, there is very little you can do to influence the future of the Java platform.

What about MS? I think that they also wanted to "embrace and extend" Java as well at some point.

My point is, that sometimes I feel like people think free software should always get its way as a matter of principle. The truth is, that you are free to write an alternative classpath. and distribute it with your own VM as it is. Quit moaning and join in with the guys at GNU Classpath [gnu.org]. Sun doesn't mind. It's focusing on its own. That's what MS did.

And this is bad why?

[SpootFinallyRegister]

Locking down the core libraries is a good thing.

Yes, people could cram more and better functionality in there, and everyone is sure their idea will just make Java better; but every new version makes Java less and less portable. Adding the same functionality in a non-core library is every bit as effective, but doesn't add the additional requirement for a new version, and will avoid breaking and deprecating enormous existing bodies of code.

If you need new or better/differently implemented functionality, you are free to add it. If Java is unable to accomodate the addition outside the standard library, then the platform has failed.

Don't get me wrong -- I'm a fan of Java. But, if the core of a language needs to be constantly updated, its an unstable language, and bad for production development. I think Java can be stable, but it requires that the developers do just what they are. Keep the standard libraries standard, and not full of every shoehorned-in functionality they can think of. Its already bloated enough.

Universally true

[pauljlucas]

Many Java developers are frustrated by the not-so-open process to improve/correct/augment the Java core libraries. Unless you work for Sun or belong to a JSR expert group, there is very little you can do to influence the future of the Java platform.

Huh? This is univerally true for any language: unless you're on the ISO C committee, for example, you have very little influence of the future of the C programming language. Ditto for C++, SQL, XQuery, and every other language with a standard. If you want to have an influence, join a committee!

Re:Universally true

[michaelggreer]

They aren't talking about the language specification, but the core libraries that ship with the JVM. They are interested in improving and augmenting the core implementation classes. However, I happen to think building a hedge around Java is the right thing. Since you can always ship these extra libraries with your product, the need for including these v. managing the stability and interoperability of Java just isn't compelling enough.

Re:Universally true

[pauljlucas]

They aren't talking about the language specification, but the core libraries that ship with the JVM.

Yes I know. It's irrelevant. STL (part of the C++ standard) can't be changed by anybody except the ISO C++ committee. The POSIX API can't be changed by anybody except it's committee. These people who want to change the Java core libraries are asking for special treatment and crying because they don't have it.

Re:Universally true

[michaelggreer]

Good point. I'm not sure, however, "crying" describes what they are doing. I don't think getting on the committee is easy, and even then its hard to make changes. Apache (a member) has not found it easy to bring in additions: hence our two logging systems.

Re:Universally true

[DrEasy]

There's a lot of wheel reinvention going on with Sun and their Java libraries. Witness the Collection Frameworks, where we already had the Java Generic Library (JGL), same deal with JAXP going against many good XML parsers. Now we have indeed two logging systems, and I'm willing to bet that we'll have some sort of JUnit copycat soon enough.

OTOH, my guess is that having Sun provide such libraries makes life easier license-wise, but IANAL.

Re:Universally true

[DrVxD]

> unless you're on the ISO C committee, for example, you have very little influence of the future of the C programming language

This is a little misleading. I sat on the ISO C++ committee's library working group for quite a while (and will resume doing so when circumstances permit), and we were always prepared to consider well thought-out proposals for library extentions (go check out Boost [boost.org] for several examples). All ISO committees have a route for interested parties to make changes to a standard, altho

cern.colt

[rowanxmas]

if you program in Java and need to do matrix stuff or, want fast data structures check out:
the CERN Colt Libraries [home.cern.ch]

Perplexed

[dark404]

I've never really understood why people get their panties in a wad over the java core library. The java core is a wealth of good tools, but the language is not limited to just those tools. If you write a good library, people will use it. If you write a crappy library, getting it included in the core java library won't make people use it, it'll just waste space. I personally would rather the core library be slower to include things until they are well tested, than to include lots of new things just because they're popular. Just because YOU write a library to calculating profit given two arguments and a couple ?'s doesn't mean *I* want it included as part of the java core.

It's cute, but it won't fly...

[dkf]

What we have here is a library that comes in several bits. The numeric and units handling parts are all very neat, but rather special-purpose (I just don't need them at all in my Java programs), and the stuff to allocate things on the stack for improved realtime performance is rather risky for non-experts to use (e.g. stack space tends to be much smaller than heap space, so you have to be careful with how much stuff you actually allocate on the stack). The RT bits also look (from a casual glance) like they'd need an update to the JVM and bytecodes to actually work...

JVM and bytecode updates are painful (can you say "Porting and Deployment" without winceing?) so those bits are a non-starter for general use, and the rest of it could (theoretically; I don't know how much it depends on the parts I've just rejected out of hand) just as happily be in a third-party lib, and they're something that Java's really good at.